Skip to content

Not yet independently audited

This page was written from verified sources but has not yet passed the independent citation audit. Treat its claims as provisional.

Guideline registry

Master table

Body Year Region Scope Citation / verified URL Status
NICE 2017; updated 2020 UK Recognition and treatment of eating disorders NG69 recommendations Current; exceptional surveillance 2024 retained
American Psychiatric Association 2023 US Treatment of eating disorders Crone C, et al. Am J Psychiatry. 2023;180:167-171. PMID 36722117 (indexed summary of the guideline); full text at psychiatryonline Current
RANZCP 2014 Australia/New Zealand DSM-5 feeding/eating disorders RANZCP guideline PDF Published guideline; currency should be monitored
Society for Adolescent Health and Medicine 2022 North America Restrictive eating disorders in adolescents/young adults J Adolesc Health. PMID 36058805 Current position paper
NICE CG9 2004 UK Eating disorders in people over 8 CG9 Superseded by → NG69
APA 2006 US Eating disorders Archived guideline PDF Superseded by → APA 2023
APA companion tool 2024 US Practice assessment tool for eating-disorder care Fochtmann LJ, et al. Focus (Am Psychiatr Publ). 2024;22:350-368. PMID 38988459 Current companion to the 2023 guideline
RANZCP (published version) 2014 Australia/New Zealand DSM-5 feeding and eating disorders; NHMRC methodology Hay P, et al. Aust N Z J Psychiatry. 2014;48:977-1008. PMID 25351912 Current in publication record; currency should be monitored
Canadian practice guidelines (children/adolescents) 2020 Canada Treatment of children and adolescents with eating disorders; GRADE Couturier J, et al. J Eat Disord. 2020;8:4. PMID 32021688 Current
Canadian transitions guideline 2025 Canada Paediatric-to-adult transition for eating disorders and mental-health conditions; GRADE + modified Delphi Dimitropoulos G, et al. J Eat Disord. 2025;13:158. PMID 40722124 Current
German S3 (revised) 2019 revision of 2011 original Germany Diagnosis and treatment of eating disorders Revision described in Resmark G, et al. J Clin Med. 2019;8:153. PMID 30700054 Revision supersedes → German S3 2011
ANZAED practice and training standards 2020 Australia/New Zealand Workforce competence: 8 treatment principles, 7 clinical practice standards, plus discipline-specific standards Heruc G, et al. J Eat Disord. 2020;8:63 (PMID 33292546); Hurst K, et al. 2020;8:58 (PMID 33292542); Heruc G, et al. 2020;8:77 (PMID 33317617) Current; consensus, not evidence-graded
USPSTF 2022 US Screening for eating disorders in adolescents and adults Davidson KW, et al. JAMA. 2022;327:1061-1067. PMID 35289876; evidence report Feltner C, et al. JAMA. 2022;327:1068-1082 (PMID 35289875) Current — I statement (evidence insufficient)
ASPEN refeeding-syndrome consensus 2020 International (US-led) Identification, risk stratification and management of refeeding syndrome in adults and children da Silva JSV, et al. Nutr Clin Pract. 2020;35:178-195. PMID 32115791 Current; not AN-specific but the operative definition for AN refeeding research
AED "Nine Truths About Eating Disorders" 2015; scientific review 2017 International Public-facing consensus statement; translated into 30+ languages Evidence review: Schaumberg K, et al. Eur Eat Disord Rev. 2017;25:432-450. PMID 28967161 Current; advocacy consensus, not a clinical guideline

URL verification status, re-checked by the independent auditor on 2026-09-02. NICE NG69 recommendations (HTTP 200), NICE CG9 (200) and the RANZCP guideline PDF (200) resolved directly. Both psychiatryonline.org URLs returned HTTP 403 to an automated request — this is bot filtering rather than evidence that the documents have moved, but it means the APA documents could not be independently confirmed live from this session by URL. The APA 2023 guideline is instead anchored to an indexed PubMed record (PMID 36722117), which was verified live, and a 2024 companion practice-assessment tool (PMID 38988459) was located in the same search. This registry catalogs documents; recommendation synthesis is in wiki/guidelines.md.

NICE NG69

Recommendation text retrieved and read during the independent audit on 2026-09-02; the following are verified against that text rather than summarized from memory. NICE recommends AN-focused family therapy (FT‑AN) for children and young people, as single-family or combined single- and multi-family therapy, typically 18–20 sessions (1.3.10–1.3.11). For adults it offers a choice of CBT‑ED, MANTRA or SSCM (1.3.4), with focal psychodynamic therapy or an untried one of the three if the first is unacceptable, contraindicated or ineffective (1.3.8). Medication must not be offered as sole treatment (1.3.24). Single measures such as BMI or duration of illness must not determine whether treatment is offered (1.2.8), and no absolute weight or BMI threshold should govern admission to day-patient or inpatient care (1.11.2); acute medical care including emergency admission is indicated for severe electrolyte imbalance, severe malnutrition, severe dehydration or signs of incipient organ failure (1.10.3). Section 1.11 warns that a person may become institutionalised by a long admission and that lack of change may indicate inpatient treatment is harmful; section 1.12 sets out the Mental Health Act 1983 framework for compulsory treatment. The 2024 exceptional surveillance review acknowledged a gap for longstanding AN but did not update the guideline.

APA 2023

APA recommends eating-disorder-focused psychotherapy for adults with AN, including normalization of eating/weight-control behaviour, weight restoration and psychological aspects. The guideline integrates assessment, treatment planning, nutrition and medical/psychiatric management. Its US service assumptions differ from NICE’s NHS pathway. The guideline is summarized in a peer-reviewed indexed article (Crone 2023, PMID 36722117) and accompanied by a 2024 practice-assessment tool (Fochtmann 2024, PMID 38988459).

RANZCP 2014

PDF retrieved and read during the independent audit on 2026-09-02. The guideline is organized around DSM-5 eating-disorder diagnostic groups — its comparative clinical-features table includes atypical anorexia nervosa alongside AN, BN and BED, with severity specifiers — and grades recommendations on the Australian NHMRC levels of evidence (EBR I = systematic review of level II studies; EBR II = randomised controlled trial). Its treatment table assigns family therapy and individual psychotherapy for AN to Level I, with antipsychotic medication graded separately for weight gain. References are accessed through late 2013. Its age means newer refeeding, adult psychotherapy and atypical-AN evidence require supplementation rather than silent application as current evidence.

SAHM 2022

The position paper emphasizes that restrictive eating disorders can be medically serious in adolescents and young adults across weights; it addresses hospitalization factors, multidisciplinary management and nutritional rehabilitation (PMID 36058805).

Canada 2020 (children and adolescents)

Developed with systematic review, GRADE and a multi-stakeholder national panel. Only two recommendations reached strong: family-based treatment, and use of the least intensive treatment environment. Weak positive recommendations covered multi-family therapy, cognitive behavioural therapy, adolescent-focused psychotherapy, adjunctive yoga and atypical antipsychotics (Couturier 2020, PMID 32021688). The last of these is the sharpest divergence in this registry: NICE states medication must not be the sole treatment and does not endorse adjunctive antipsychotics, while the Canadian panel issued a weak positive recommendation from substantially the same trial evidence — and a systematic scoping review finds seven randomized olanzapine trials, two quetiapine, one risperidone and none for aripiprazole (Thorey 2023, PMID 36928656).

Canada 2025 (transitions)

The only guideline located in this registry that addresses the paediatric-to-adult transition. Built on a scoping review screening 14,350 records with 419 studies included (199 primary), it issued strong recommendations for integrated, collaborative transition approaches involving young people, families and providers, and for the Transition Readiness Assessment Questionnaire, while stating that certainty of evidence for specific interventions and tools was generally low and that only three studies contributed key-outcome data (Dimitropoulos 2025, PMID 40722124).

RANZCP 2014 — published version

The indexed publication records that for AN the guideline recommends treatment as an outpatient or day patient in most instances (the least restrictive environment), hospital admission for those at medical or psychological risk, a multi-axial collaborative approach, family-based therapies in younger people, specialist therapist-led manualized psychological therapies in all age groups with longer-term follow-up, and a harm-minimisation approach in chronic AN. It states that no specific treatment can be recommended for ARFID because no trials exist (Hay 2014, PMID 25351912). The harm-minimisation position and the all-ages manualized-therapy recommendation are both points of difference from NICE.

ANZAED practice and training standards (2020)

The only document in this registry that specifies who is competent to deliver eating-disorder treatment rather than what should be delivered. Developed through four consultation stages including a workshop of roughly 100 clinicians and an open online process. Eight treatment principles: early intervention is essential; co-ordination of services is fundamental; services must be evidence-based; involvement of significant others is highly desirable; personalised treatment is required; education/psychoeducation is included in all interventions; multidisciplinary care is required; a skilled workforce is necessary. Seven clinical practice standards: diagnosis and assessment; the multidisciplinary team; therapeutic alliance; knowledge of evidence-based treatment; knowledge of levels of care; relapse prevention; professional responsibility (Heruc 2020, PMID 33292546), with separate mental-health-professional (PMID 33292542) and dietitian (PMID 33317617) standards. Consensus documents; no evidence grading.

USPSTF 2022 — a formal "insufficient evidence" finding

The Task Force commissioned a systematic review of screening for eating disorders in adolescents and adults with normal or high BMI, explicitly excluding people who are underweight or already showing physical signs. Across 57 studies (N=10,773), no study directly evaluated the benefits and harms of screening, and none of 40 intervention RCTs enrolled a screen-detected population; SCOFF at cut point ≥2 had pooled sensitivity 84% (95% CI 74–90%) and specificity 80% (65–89%) in adults across 10 studies (n=3,684) (Feltner 2022, PMID 35289875). The resulting recommendation is an I statement — evidence insufficient to assess the balance of benefits and harms (Davidson 2022, PMID 35289876). No other guideline in this registry addresses population screening directly.

ASPEN refeeding-syndrome consensus (2020)

Proposes a clinical definition and risk stratification: refeeding syndrome is a decrease in any one, two or three of serum phosphorus, potassium and/or magnesium by 10–20% (mild), 20–30% (moderate) or >30% and/or organ dysfunction attributable to those falls or to thiamine deficiency (severe), occurring within 5 days of reintroducing calories (da Silva 2020, PMID 32115791). Applying these criteria retrospectively to 35 observational studies did not reconcile a reported incidence range of 0% to 62%, and the reviewers judged pooled incidence measures meaningless (Cioffi 2021, PMID 34134001). No AN-specific guideline in this registry adopts the ASPEN definition explicitly.

International concordance

A systematic comparison of nine evidence-based eating-disorder guidelines found that they "endorse the main empirically validated treatment approaches with considerable agreement, but additional recommendations are largely inconsistent", and noted that people with eating disorders frequently do not receive an evidence-based treatment at all (Hilbert 2017, PMID 28777107). Guideline concordance is therefore not the binding constraint on care quality.

Disagreements and gaps

Topic Agreement Difference/gap
Psychotherapy Specialist ED-focused care Named menus and evidence grades differ
Adolescents Family/caregiver involvement Delivery variants and contraindications under-studied
Medication Not sole treatment Olanzapine positioning varies
Admission Multivariable risk Threshold tables and service capacity differ
Refeeding Monitor electrolytes/physiology Starting energy and prophylactic phosphate vary; none specifies weight-indexed dosing despite randomized evidence of underfeeding in atypical AN (Garber 2024, PMID 38179719)
Severe enduring AN Individualized care No validated futility or terminal criteria
Compulsion Legal safeguards Direct outcome evidence sparse; law jurisdiction-specific
Atypical AN Increasing recognition Treatment evidence and operational severity remain limited
Adjunctive antipsychotics All bodies agree medication is not sole treatment Canada issues a weak positive recommendation for atypical antipsychotics; NICE does not endorse them adjunctively; aripiprazole appears in some international guidance on zero randomized trials (Couturier 2020, PMID 32021688; Thorey 2023, PMID 36928656)
Population screening USPSTF finds evidence insufficient (I statement); no other guideline addresses screening directly (Davidson 2022, PMID 35289876)
Care transitions Only one guideline worldwide addresses paediatric-to-adult transition, at generally low certainty (Dimitropoulos 2025, PMID 40722124)
Workforce competence Only ANZAED specifies training standards; no evidence grading attached (Heruc 2020, PMID 33292546)
Harm minimisation in chronic AN RANZCP recommends it explicitly; NICE addresses longstanding AN only through its 2024 surveillance gap statement (Hay 2014, PMID 25351912)
Refeeding-syndrome definition Monitoring is universal Only the non-AN-specific ASPEN consensus provides an operational definition, and applying it does not reconcile incidence estimates (da Silva 2020, PMID 32115791; Cioffi 2021, PMID 34134001)

Watch list

  • NICE surveillance/update status after 2024 exceptional review.
  • APA corrections or focused updates to the 2023 guideline.
  • RANZCP replacement or formal currency statement.
  • Updated pediatric/adolescent medical instability and refeeding positions.
  • Guidance explicitly addressing atypical AN, gender diversity and higher-weight patients.
  • Added by the audit (2026-09-02). Refeeding guidance expressed in kcal/day rather than kcal/kg: randomized data show fixed-calorie protocols underfeed patients with atypical AN by roughly 25% per kilogram and delay their medical stabilization (Garber 2024, PMID 38179719). No current guideline in this registry addresses weight-indexed dosing.
  • Added by the audit (2026-09-02). Bone care: population-scale fracture risk is now quantified (osteoporotic-fracture HR 7.50, 95% CI 5.62–10.01) while no agent is approved and no trial uses fracture as an endpoint (Cyrenne-Dussault 2026, PMID 41109616; Robinson 2017, PMID 28554377). Guidance on when to treat bone loss remains unsupported in either direction.
  • Added by the deepening pass (2026-09-02). Whether any guideline adopts the ASPEN refeeding-syndrome definition, which would make incidence figures comparable across studies for the first time (da Silva 2020, PMID 32115791).
  • Added by the deepening pass (2026-09-02). Whether any body revisits adjunctive antipsychotic recommendations in light of the scoping review showing zero randomized aripiprazole trials (Thorey 2023, PMID 36928656).
  • Added by the deepening pass (2026-09-02). Whether guidelines begin to address maladaptive exercise assessment, now measured at 88% lifetime prevalence across a 31,671-person multicountry sample with prevalence varying three-fold by instrument (Watson 2026, PMID 41115789).
  • Added by the audit (2026-09-02). Compulsory treatment: the first large register cohort of post-compulsion mortality has been published (Bager 2026, PMID 42383339) without a comparator pathway; watch for whether any guideline body treats it as outcome evidence, which it is not.